双置信对比解码用于检索增强生成
Dual-Confidence Contrastive Decoding for Retrieval-Augmented Generation
AI总结：
针对多文档检索增强生成中的内部证据冲突问题，提出无需训练的双置信对比解码方法，结合文档级和词元级置信度选择正负流，在DRQA等基准上取得最佳平均性能。
摘要：
检索增强生成（RAG）日益要求模型从多个检索到的文档中回答问题，其中只有部分来源相关，且检索到的文档集可能包含过时、噪声或矛盾的证据。现有的对比解码方法主要关注模型内部记忆与检索上下文之间的冲突。相比之下，我们研究了多文档RAG中上下文内部冲突这一互补问题。为了评估这一设置，我们引入了DRQA，一个基于企业深度研究场景的事实冲突问答基准，其中的答案基于合成企业特定事实，这些事实设计为无法从模型内部记忆中恢复。我们进一步提出了双置信对比解码（DCCD），一种无需训练的解码方法，结合了文档级置信度（估计文档是否足以回答问题）和词元级置信度（估计该文档是否支持置信的下一个词元预测）。DCCD使用这些双置信信号选择正负文档条件流，并通过其置信度差距缩放文档级对比。在DRQA和标准多文档问答基准上，DCCD在全上下文和对比解码基线中取得了最佳平均性能，在DRQA上提升最大。这些结果强调了当检索到的证据内部存在冲突时，基于来源感知、置信门控解码的重要性。
英文摘要：
Retrieval-augmented generation (RAG) increasingly requires models to answer questions from multiple retrieved documents, where only some sources are relevant and the retrieved bundle may contain stale, noisy, or conflicting evidence. Existing contrastive decoding methods primarily focus on resolving conflicts between the model's internal memory and the retrieved context. In contrast, we study the complementary problem of intra-context conflict in multi-document RAG. To evaluate this setting, we introduce DRQA, a factual-conflict question answering benchmark derived from enterprise deep-research scenarios, where answers are grounded in synthetic enterprise-specific facts that are designed not to be recoverable from the model's internal memory. We further propose Dual-Confidence Contrastive Decoding (DCCD), a training-free decoding method that combines document-level confidence, which estimates whether a document appears sufficient for answering the question, with token-level confidence, which estimates whether that document supports a confident next-token prediction. DCCD selects positive and negative document-conditioned streams using these dual-confidence signals and scales a document-level contrast by their confidence margin. Across DRQA and standard multi-document QA benchmarks, DCCD achieves the best average performance among full-context and contrastive decoding baselines, with the largest gains on DRQA. These results highlight the importance of source-aware, confidence-gated decoding when retrieved evidence is internally conflicting.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
